Zededa and Submer have partnered to ship modular, liquid-cooled edge AI infrastructure for high-density GPU inference in places with out conventional knowledge facilities.
Linking Submer’s liquid-cooled AI infrastructure with Zededa’s edge intelligence platform, the companions will ship scalable, safe and resilient edge AI deployments.
“AI is quickly shifting from centralized cloud environments into real-world operations, from industrial websites to telecom networks and distant power infrastructure,” says Patrick Smets, CEO of Submer. “Delivering that intelligence requires purpose-built AI infrastructure that operates effectively in environments the place conventional knowledge facilities merely can’t exist. By combining Submer’s liquid-cooled high-density AI infrastructure with Zededa’s edge intelligence platform, we’re enabling organizations to deploy scalable, resilient AI infrastructure wherever it’s wanted.”
Three modular solutions are provided: edge pods (from 2 to eight GPUs); ruggedized micro-data facilities (to 168 GPUs), and; megawatt-scale containerized methods (as much as 800 GPUs).
Submer’s liquid cooling expertise improves power effectivity, reduces water consumption and permits high-density GPUs in harsh environments.
Zededa’s software-defined resilience reduces {hardware} redundancy prices by redistributing workloads throughout node failures, guaranteeing uptime and reducing whole price of possession.
The deployments tackle a variety of AI workloads reminiscent of real-time pc imaginative and prescient, predictive upkeep and industrial automation in order that AI can run on the edge in distant or excessive environments.
The primary pilot deployments are anticipated to start with industrial and telecommunications clients later this yr.
The companions will deal with delivering the rising edge AI infrastructure essential to convey workloads from centralized cloud environments to operational and industrial settings.
Late final yr Zededa rolled out full-stack edge Kubernetes to deal with large-scale AI deployments.
Associated
AI infrastructure | knowledge facilities | edge AI | edge infrastructure
